01-09-2016 | By Paul Whytock
A collaboration on a 28nm embedded flash (eFlash) process for making microcontrollers (MCUs) aimed at green and autonomous vehicles applications has been agreed between pure-play foundry TSMC and Renesas Electronics.
These companies have worked together on MCUs with on-chip flash memory since the 90nm technology generation. Four years after working together on 40nm MCU platform and production they are now extending their collaboration to develop 28nm MCUs.
Through this collaboration, Renesas’ Metal-Oxide-Nitride-Oxide-Silicon (MONOS) eFlash technology will combine with TSMC’s low-power 28nm high-K metal gate process technology. MONOS is a structure in which each transistor in the flash cell consists of three layers, oxide, nitride, and oxide on a silicon base with a metal control gate at the top.
The projected MCUs are expected to find applications such as autonomous vehicle sensor control, coordinated control among electronic control units (ECUs), fuel-efficient engine control for green vehicles and efficient motor inverter control for electric vehicles.
Next-generation MCUs can fulfil the high performance and demanding safety requirements of future autonomous-driving cars since they enable high precision sensing using 3D radar to monitor the environment surrounding the vehicle, integration of data from multiple sensors, as well as real-time judgment processing for autonomous operation.
ECUs capable of safely controlling autonomous-driving functionality require next-generation control MCUs that contribute to fast processing of complex control tasks (including fail-operational capabilities, security, and support for coordinated control among multiple ECUs), power efficiency of the overall system, and functional safety.
To meet increasingly stringent emission regulations, fuel-efficient engines for next-generation green vehicles require more computing performance to implement new combustion systems as well as adequate on-chip flash memory to accommodate larger firmware programs.
There is also now a demand for MCUs with improved computing performance and greater function integration to enable more efficient and compact motor inverters. There is also a need for large-capacity flash memory to facilitate greater flexibility for the environmental regulations and standards of various countries as well as to enable over-the-air (OTA) wireless updating of control programs.
The collaborating companies say that with the 28nm eFlash process technology developed through their work MCUs can meet the demands of next-generation automotive computing by delivering a maximum of more than four times the program memory capacity and greater than four-fold performance improvement compared to the current 40nm technology. Other enhancements on the new MCUs include the use of multiple CPU cores, more advanced security, and support for multiple interface standards.
The first automotive MCUs employing this new 28nm process technology are planned for sample shipment in 2017.